Oseltamivir is effective for influenza treatment if started within 2 days after symptom onset, but little is known regarding its effects on illness severity and household transmission when initiated later in the course of illness. The aims of this research are to 1) assess the efficacy of early (<48 hours after onset) and late (48 to 143 hours) initiation of oseltamivir therapy for reducing duration of influenza symptoms, illness severity, and duration of viral shedding;2) understand how early and late therapy of an index case influences the transmission dynamics of influenza in households;3) determine if prior influenza vaccination alters the response to antiviral therapy;and 4) determine if the incidence of secondary complications can be reduced by late initiation of antiviral therapy. This will be a double-blind, placebo-controlled clinical efficacy trial within a defined population cohort. Patients 12 months through 79 years of age with medicallyattended acute respiratory illness <7 days duration will be eligible. Potential participants will be identified through active surveillance in primary care clinics and a hospital. Nasal swabs will be obtained for rapid antigen test from consenting patients. Reverse transcription polymerase chain reaction (PCR) will be performed if the enrollment swab is negative by rapid antigen test. Participants with influenza will be randomized to receive oseltamivir or placebo in a 2:1 ratio. Hospital inpatients with influenza will not be randomized to receive placebo, and they will be analyzed as a separate group to compare those initiating oseltamivir treatment early vs. late. The target sample size is 525 participants with laboratory-confirmed influenza. Participants with influenza will record temperature and symptoms twice each day until symptoms resolve. They will be swabbed again on day 5 or 6, and viral culture will be performed to identify those who are still shedding. Household transmission will be assessed in a subgroup, and consenting household members will be tested for influenza if they develop acute respiratory illness. Daily samples for viral culture will also be collected from a subset of 75 patients to assess duration of shedding. Analyses will compare illness duration, viral shedding duration, severity, and household transmission in patients initiating antiviral treatment early and late relative to placebo. The impact of influenza vaccination on these parameters will be assessed.